| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This lexer is being written as a chunklet, not quite quite complete yet
as it lacks some functionality to make it generally useful for things,
but good enough for the gluegen use case now. So it's in the repo now
and we can go ahead and use it for this instead of having this
hacked-to-pieces third party thing that's nowhere near as efficient. In
the future the goal is to have a decently usable library for any sort of
C metaprogramming needs, not just within this project.
This design does the data-oriented thing of storing as little about each
token as possible (just 5 bytes), and re-lexing specific pieces only
when necessary. In some cases, full validation of correct syntax is
only possible through this secondary step. Since the use case is
metaprogramming and code-generation rather than the reimplementation of
Clang, this slight sloppiness in validation doesn't seem too bad.
It's been a while since I did a proper performance measurement of this
code but in some crude tests I did in the past the primary tokenisation
step was running at well over 500MB/s, which is fast enough for me. It's
possible that this version is a tiny bit slower due to the added
complexity of making UCNs work. UCNs, incidentally, are one of the
dumbest features of C by far. However, unlike trigraphs - which this
lexer does not handle - UCNs are still in the language, so it's kind of
sort of necessary to support them.
It's probably still possible to come up with a faster design with some
SIMD trickery, but the main loop is currently branchless and the lookup
tables are too big for the SIMD lookup things, so that seems kind of
hard. A separate SIMD path for whitespace or comment runs seems dubious
as it would introduce branch mispredictions everywhere. I also made
previous attempts to unroll the main loop and every attempt just made it
slower, so I guess code size is a significant factor.
Optimising the secondary tokenisation of identifiers (and later numerals
and string/character literals once those are handled) is still on the
cards, but since that happens less often, I don't know how much
difference it'll make.
At any rate, in a multithreaded context this thing would already come
pretty close to SSD speeds, if open-read-close syscall overhead doesn't
get in the way first. I imagine it's fast enough for anyone who hasn't
*already* written something faster.
|
| |
|
|
|
| |
Saves multiple kilobytes, somehow. Unlikely to make much of a
performance difference since it's mostly init/shutdown stuff.
|
| |
|
|
| |
Quick unused header cleanup pass while we're at it.
|
| |
|
|
|
|
|
| |
Turns out, there's no need for the trailing underscores. Plus, glibc
does some stupid stuff with __attribute__ for non-GCC compilers. Not
that that matters here, but it seems like a good practice just to use
the forms that never have such problems. And it's shorter too.
|
| |
|
|
|
|
|
|
|
|
| |
I had at one point tried to macrofy the repetitive functions. After
being talked down from that madness, I LSP-expanded all the macros out
and went through rewriting the crazy expressions into constants.
Except those two for some reason, because I'm blind I guess.
Thanks bill for pointing this out.
|
| |
|
|
|
|
|
|
| |
They're legally unnecessary as far as I know, and kind of annoying to
maintain on a long-term basis.
This was done with the consent of all 3 other contributors, in case
anyone was wondering.
|
| | |
|
| |
|
|
|
|
|
|
| |
- Add reverse-put functions for stack-based and/or left-pad-inspired
use cases.
- Fix an incorrect return value.
- Clean up the doc comments, fixing a few errors and omissions.
- Make some trivial functions static inline.
|
| |
|
|
|
|
|
|
|
| |
My new programming style is branch hints. All non-confusing branches
must be hinted when I can be bothered. It's faster, sometimes, maybe.
Also, start trying to use more signed sizes in at least some of the
places where it makes sense. Unsigned sizes are surprisingly
error-prone!
|
| |
|
|
|
| |
Reminder not to actually use any of the code I write until it's at least
been included in a few releases of something. :^)
|
| |
|
|
|
|
|
|
|
|
|
| |
Nothing really works yet, but at least test.h and fastspin are fixed and
some of the issues with RTTI and libdl and stuff are maybe kind of
sorted, subject to more testing later.
The main issue now seems to be the cvar interface not quite lining up
and crashing pretty much immediately. That'll probably take a lot more
debugging to figure out, which likely still won't be a priority for
quite a while.
|
|
|
A lot of this is random WIP from a while back, at least a month ago, and
is being committed now to get it out of the way so that other patches
can be brought in and integrated against it without causing headaches.
Also rolled into this commit is a way to distinguish plugin_unload from
exiting the game. This is required for another soon-to-be-integrated
feature to avoid crashing on exit, and could in theory also be used to
speed up unloading on exit in future. While we're at it, this also
avoids the need to linearly scan through the plugin list to do the
old branch unloading fix, because we can.
Rough summary of the other smaller stuff I can remember doing:
- Rework bitbuf a bit
- Add some cryptographic nonsense in ac.c (not final at all)
- Introduce the first couple of "chunklets" libraries as a sort-of
subproject of this one
- Tidy up random small bits and bobs
- Add source for a small keypair generation tool
- Rework democustom to be very marginally more useful
|
Michael Smith
